Special Report – Mule Creek State Prison Must Improve Its Oversight Of Some Employees’ Work Hours and Timekeeping

Special RepoRt Mule cReek State pRiSon MuSt iMpR ove itS oveRSight of SoMe eMployeeS’ WoRk houRS and tiMekeeping office of the inSpectoR geneRal State of California april 2011 Contents Executive Summary .................................................................................................................... 1 Introduction ................................................................................................................................. 6 Objectives, Scope, and Methodology ......................................................................................... 9 Finding 1 ................................................................................................................................... 10 Many of Mule Creek State Prison’s Mental Health Employees Received Full-Time Pay, But Appeared to Work Only Part Time Recommendations ......................................................................................................... 21 Finding 2 ................................................................................................................................... 23 Mule Creek State Prison’s Academic Teachers, Vocational Instructors, and Educational Supervisors Appeared to Work Less Than Full Days Recommendations ......................................................................................................... 27 Finding 3 ................................................................................................................................... 29 Ineffective Supervisory Oversight and Personnel Practices Concerning Employee Timesheets at Mule Creek State Prison Have Resulted in Costly Mistakes Recommendations ......................................................................................................... 36 Appendix ................................................................................................................................... 37 California Department of Corrections and Rehabilitation’s Response ..................................... 40 Executive Summary The Office of the Inspector General Findings in Brief initiated a special review regarding employee timekeeping and workload at The Office of the Inspector General found that many of Mule Creek State Prison’s (Mule Creek) mental health and Mule Creek State Prison (Mule Creek). educational employees had large numbers of unaccounted-for Mule Creek is currently the only prison hours primarily because they received full pay; but according to in the state with an electronic security the prison’s electronic security system data, those employees appeared to work only part time inside the prison’s secured system that records the identity of each perimeter. In total, these employees’ unaccounted-for hours cost employee who enters and exits the about $272,900 during a three-month period (or, at this rate, prison’s secured perimeter—the area of nearly $1.1 million in a year). Our data analysis for the period of June through August 2010 revealed the following: the prison within the electrified fence—as • Eleven of the 13 psychiatrists averaged working the well as the date and time of each entry equivalent of about 26 to 34 hours per week. On average, and exit. The warden installed this system psychiatrists receive an annual salary of $245,000. for security purposes to determine which • Twenty-six of the 31 psychologists averaged working the of Mule Creek’s employees are inside the equivalent of about 28 to 39 hours per week. On average, psychologists receive an annual salary of $103,000. secured perimeter during an emergency. We used the system to determine how • All seven licensed clinical social workers averaged working the equivalent of about 28 to 38 hours per week. On long employees who are supposed to work average, licensed clinical social workers receive an annual inside the secured perimeter were actually salary of $80,000. present inside the secured perimeter. We • All 12 academic teachers and five vocational instructors analyzed the system’s records from a averaged working the equivalent of about 33 to 39 hours per week. On average, these educators receive an annual salary three-month period ending August 2010. of $77,000. • The principal and the two vice principals averaged working We compared the number of hours the the equivalent of about 33 to 35 hours per week. On employees spent inside the secured average, these three employees receive an annual salary of $89,000. perimeter to hours captured by other personnel-related data systems, which In addition, we found that Mule Creek could increase its clinicians’ available mental health patient hours by 25 percent if record hours the employees were paid, it switches their alternative work schedules to traditional eight- hours they spent in training, and hours hour shifts. they took as time off. We found that a Finally, we found that timekeeping mistakes on a sample of wide range of the prison’s employees employee timesheets resulted in employees being over- or had unaccounted-for hours: that is, hours undercharged more than $108,000 in leave hours. for which they were paid but which they did not spend inside the secured perimeter, in training, or in time off. We focused our review on mental health and educational employees because these two groups had relatively large numbers of unaccounted-for hours at a high financial cost. The mental health employees, consisting of psychiatrists, psychologists, and licensed clinical social workers, work under the direction of the prison’s chief executive officer. The chief executive officer has a dual reporting relationship to the California Department of Corrections and Rehabilitation’s (CDCR) chief deputy secretary, Division of Correctional Health Care Services for mental health and dental program issues, and to the federally appointed receiver for medical service delivery issues. The educational employees, consisting of academic teachers, vocational instructors, and their supervisors, work for the prison’s warden and CDCR. State of California • April 2011 Page 1 The three-month cost of these two groups’ unaccounted-for hours totaled $272,900; at this rate, the cost comes to nearly $1.1 million in a year. Based on our discussions with the prison’s chief executive officer—who alluded to similar problems at another prison—we are concerned that other prisons may also have employees with large numbers of unaccounted-for hours, but because the remaining 32 state-run prisons do not have similar electronic time-recording security systems, it is unknown whether the problems identified at Mule Creek constitute an isolated occurrence at one prison. When analyzing the number of hours Mule Creek’s mental health employees worked inside the secured perimeter, we found that most of their unaccounted-for hours stemmed from their not averaging a full day’s work within that perimeter, where the clinicians are paid to provide mental health services to inmates. The cost of these employees’ unaccounted-for hours during the three-month period was $242,389 of the $272,900. According to the security system’s electronic data, most of Mule Creek’s psychiatrists, psychologists, and licensed clinical social workers regularly arrived to work late and left early, averaging as a group only 8.4 hours per day of their scheduled ten-hour shifts inside the secured perimeter. We followed up with several of the employees with unaccounted-for hours to obtain their perspective. Three of the employees we spoke with admitted to leaving early but offered the justification that their Memorandum of Understanding (MOU) with the state allows them to leave early after they have completed their work. Although these salaried employees’ MOU allows for flexible work hours, it does not grant permission to leave early on a daily basis. In other cases, employees we spoke with offered additional partial excuses for the hours they spent outside of the prison’s secured perimeter, explaining that they were working from home or reviewing inmate files in the administrative building. However, according to Mule Creek’s chief of mental health, working from home is not permitted. Moreover, our examination of the prison’s case records filing system demonstrated that the mental health employees in question did not review enough inmate files in the administrative building for those reviews to make much of a difference in their shortage of work hours. Employees who leave early on a regular basis— even if they complete all of their work—get paid for not working and set a poor example to their coworkers. Supervisors could instead give these employees additional assignments or responsibilities during the day to keep them fully occupied and productive or could alter their schedules to make their work time more efficient. The structure of the work schedules for most of the mental health employees may contribute to their not working a full day inside the secured perimeter. The current schedule of four ten- hour shifts per week leaves most of them with about three hours per day of non-patient or administrative time. If Mule Creek changes these employees’ work hours to a schedule of traditional eight-hour shifts, five days per week, the prison could increase its mental health patient hours by up to 25 percent while decreasing its non-patient time by 58 percent. This change could potentially limit the clinicians’ primary cause for leaving early: excessive non-patient time. Our analysis further revealed that academic teachers, vocational instructors, and educational supervisors also had unaccounted-for hours; they only worked an average of 6.6 to 7.8 hours of their scheduled eight-hour shifts inside the secured perimeter. Since all of these employees were Bureau of Audits, Office of the Inspector General Page 2 paid for full-time work, the total cost of their unaccounted-for hours is $30,511 for the three- month period; if the employees continued to leave the prison during their work time at the same rate, the total cost of their unaccounted-for hours would come to $122,044 each year. During our conversations with four academic teachers and two vocational instructors about the number of hours they typically worked, all six of these employees claimed they worked outside the prison frequently: teachers called their outside time “off-site preparation time” and instructors said they sometimes attend meetings outside of the prison. One of the employees explained that he used his off-site preparation time as part of his daily commute. However, the principal and the two vice principals at Mule Creek noted specifically that employees must formally request to use off-site preparation time or have off-site meetings, and that such requests must be approved in advance; significantly, none of them remembered receiving requests in more than a couple of instances during the period we reviewed, and none of them granted it for commuting purposes. Additionally, we noted that the principal and the two vice principals themselves worked inside the prison’s secured perimeter an average of only 6.6 to 7.0 hours of their scheduled eight-hour day, or the equivalent of only 33 to 35 hours per week. When supervisors do not hold themselves accountable, they set a poor example for employees to follow and are themselves less likely to hold subordinate employees accountable for working a full day. Finally, we found that timesheet mistakes made by employees, their supervisors, and the personnel office’s staff contributed to our difficulty in reconciling employee work hours. For example, our review of 51 employees’ timesheets over a three-month period revealed 14 of them failing to report a total of 23 full days of time off on their timesheets when they did not work. No supervisor caught any of these mistakes. At the employees’ compensation rates, this unreported time off totaled about $16,000. Furthermore, we found that once timesheets reached the prison’s personnel office, they are subject to further errors when employees who work in the personnel office tally and enter the employees’ self-reported time off into the prison’s leave-accounting system. In a sample of 325 timesheets pertaining to 112 employees, we found 73 timesheets with errors. These mistakes resulted in some employees being undercharged a total of 1,582 hours of time off (worth $86,212) and some employees being overcharged a total of 90 hours of time off (worth $6,532). In total, these errors resulted in over $108,000 in unreported time off and other timekeeping mistakes. Had we not brought these errors to Mule Creek’s attention, the undercharged employees would have been able to use the hours again or receive the cash value of those hours once they separate from state service. Conversely, the overcharged employees would not have been able to take the time off that they had earned. Recommendations: To ensure that Mule Creek State Prison receives the most productivity and value from its psychiatrists, psychologists, and licensed clinical social workers, the Office of the Inspector General recommends that the prison’s chief executive officer, in accordance with and to the extent permissible under his employees’ labor agreements, take the following actions: • Develop a method to hold supervisors and managers in mental health services accountable for ensuring that their employees complete required daily tasks and obtain prior permission before the employees leave work early. State of California • April 2011 Page 3 • Require all supervisors to maintain logs tracking the number of times employees request to leave early and, when necessary, adjust employees’ workload accordingly. • Investigate and take administrative actions against employees who leave work before completing their daily tasks or without prior permission from their supervisors. • Train all supervisors and managers on how to initiate progressive discipline on employees when appropriate. To ensure that the state receives the most productivity and value from its mental health clinicians, the Office of the Inspector General recommends that the California Department of Corrections and Rehabilitation, Division of Correctional Health Care Service (DCHCS), in accordance with the employees’ labor agreements, take the following action: • Evaluate the effectiveness of delivering mental health services to inmates using alternative work schedules for mental health clinicians at all state prisons and, in cases where DCHCS determines that alternative work schedules deliver fewer patient services or are otherwise less effective than a traditional five-day-per-week schedule, eliminate their usage. To ensure that Mule Creek State Prison receives the most productivity and value from its academic teachers, vocational instructors, and educational supervisors, the Office of the Inspector General recommends that the warden, in accordance with and to the extent permissible under his employees’ labor agreements, take the following actions: • Develop a method to hold supervisors in the prison’s education department accountable for ensuring that their employees complete their required daily tasks and obtain prior permission before they leave work early. • Require all supervisors to maintain logs tracking the number of times employees request to leave early and, when necessary, adjust employees’ workloads accordingly. • Investigate and take administrative actions against employees who leave work before completing their daily tasks or without prior permission from their supervisors. • Train educational supervisors on how to initiate progressive discipline on employees when appropriate. To ensure that all employees understand the importance and methods of accurate timekeeping, the Office of the Inspector General recommends that the warden take the following actions: • Provide training to all employees on how to properly complete timesheets. This training should include an emphasis on the importance of totaling the various types of leave hours charged each month and on writing legibly. • Provide training to all supervisors and managers on how to properly review completed timesheets. This training should include an emphasis on ensuring that employee leave hours are accurate and on returning incomplete or inaccurate timesheets to employees for corrections. Bureau of Audits, Office of the Inspector General Page 4 • Require all employees to use the prison’s form for requesting and reporting time off when taking time off from work for any reason. • Study whether Mule Creek has a sufficient number of timekeepers. If the prison lacks a sufficient number of timekeepers, redirect employees in the Personnel Office to assist the existing timekeepers. • Instruct timekeepers to reject incomplete or illegible timesheets. Where appropriate, provide remedial instruction to employees or supervisors who persist in not following existing policies related to timesheets. • Provide training to timekeepers, payroll specialists, and payroll specialist supervisors on the correct method of processing and reviewing employees’ monthly leave activity and leave balances in the California Leave Accounting System. State of California • April 2011 Page 5 Introduction Mule Creek State Prison (Mule Creek) opened in June 1987 and, as of August 2010, employed approximately 1,300 people. The prison was designed to house 1,700 inmates; yet, during 2010, Mule Creek typically housed between 3,500 and 3,800 inmates. The annual operating budget for the prison during fiscal year 2009-2010 was approximately $137 million. Mule Creek houses most of its inmates in three facilities within a perimeter, which is secured by an electrified fence (commonly referred to as the secured perimeter). Within the secured perimeter, the prison’s three facilities house inmates classified from Level III (medium-to-high security) to Level IV (maximum security). Each of these facilities houses inmates with “sensitive needs” who, because of their crime, notoriety, or gang affiliations, cannot mix with general population inmates because they may be subject to harm. Mule Creek also houses inmates in a minimum- support facility located adjacent to the secured perimeter; the Level I (minimum-security) inmates in this facility work outside of the secured perimeter and participate in the prison’s community work programs in the city of Ione. According to the California Department of Corrections and Rehabilitation’s (CDCR’s) comparative statistics, Mule Creek housed the largest total number and highest percentage of inmates with recognized mental health conditions in the state as of June 2010. More than half of Mule Creek’s inmates participate in its Enhanced Outpatient Program and Correctional Clinical Case Management System. Mule Creek also has a Mental Health Outpatient Unit and a Correctional Treatment Center, where employees provide inmates with nursing and mental health crisis care. Mule Creek has two distinct hiring authorities: the chief executive officer and the warden Mule Creek’s chief executive officer (CEO) is responsible for hiring and managing medical, dental, and mental health employees and services at the prison. The CEO has a dual reporting relationship to CDCR’s chief deputy secretary, Division of Correctional Health Care Service for mental health and dental program issues, and to the federally appointed receiver (California Prison Health Care Services) for issues related to the delivery of medical services. According to Mule Creek’s August 2010 payroll records, the CEO employed about 290 individuals. The prison’s warden, who works under the direction of CDCR, is responsible for hiring and managing all other employees and for ensuring the safety and security of the prison. For the same time period, Mule Creek’s warden employed about 990 individuals. Staff Accountability System Figure 1: Staff Accountability System Card Reader Mule Creek is currently the only prison in the state to have an electronic card swiping security system, called the Staff Accountability System (accountability system), which allows the prison to track when its employees are inside of the secured perimeter. Whenever employees enter and exit through either of the two gates leading into the prison’s secured perimeter (the front gate and the rear sally port gate), a gate officer is required to swipe Source: Office of the Inspector General. employees’ identification cards through a scanner that records the Bureau of Audits, Office of the Inspector General Page 6 precise times of their entries and exits (see photo in Figure 1 on the previous page). According to the warden, the prison implemented this system as a safety measure because it provides a quick and efficient way to account for all employees inside the prison during an emergency. The accountability system became operational on January 29, 2010, in a trial mode. However, data from the system was not retained until after May 10, 2010, when all of the prison’s employees received new identification cards and the system was fully implemented. Employees represented by various bargaining units have memorandums of understanding Employees at Mule Creek are represented by a number of collective bargaining units and Table 1: Comparison of Labor Agreement Provisions Contained in Three Bargaining Units’ Memorandums of Understanding Represented Provisions Referring to Provisions Referring to Management Employees Electronic Monitoring/ Rights to Monitor Work and Unit We Discuss Timekeeping Set Work Hours 3 Education If an employee believes that the Management determines … products, Personnel State’s use of current or future services, and standards which must be technology is being used for the met by Fair Labor Standards Act (FLSA) purpose of harassment he/she excluded employees. FLSA exempt/excluded may grieve such action under employees may be required to work specific Article 6. (21.2) hours to provide services when deemed necessary by management. (19.1.3(F)(1)) 16 Psychiatrists For the purposes of regular pay Management can require Unit 16 (salary), the work of Bargaining employees to work specified hours. Subject Unit 16 employees cannot be to prior notification and management standardized in relation to a concurrence, Unit 16 employees have the given period of time. As in the flexibility to alter their daily and weekly past, the State reserves the right work schedules. Unit 16 employees are to adequately assess the hours responsible for keeping management worked by Unit 16 employees. apprised of their schedule and must receive No time clock or time keeping prior approval from management for the device shall be implemented. use of accrued leave for absences of any (7.6(B)) duration. (7.6(C)(4)) 19 Psychologists Within the parameters established Management determines the products, and Licensed by management, the employee services, and standards which must be met Clinical Social shall be given the flexibility in by FLSA-exempt employees. FLSA exempt Workers determining how and when work employees are expected to work the hours is done, provided assigned duties necessary to accomplish their assignments are performed satisfactorily. or fulfill their responsibility. Their work load The quality of work performed, will normally average 40 hours per week the work product itself, and over a 12 month period. However, inher- the fulfillment of professional ent in their job is the responsibility and duties should be the focus of expectation that work weeks of longer du- the evaluation. If an employee ration may be necessary. Management can fails to fulfill this function, it may require FLSA-exempt employees to work indicate the need for a more specified hours. However, subject to prior fixed schedule in terms of being notification and approval, FLSA-exempt available. No time keeping device employees have the flexibility to alter their shall be implemented. (6.1(B)(6)) daily and weekly schedules. (6.1(B)(2,5,6)) Source: Employee Memorandums of Understanding. State of California • April 2011 Page 7 have contractual labor agreements with the state. These labor agreements, referred to as Memorandums of Understanding (MOUs), are negotiated and agreed to by the employees’ unions and by representatives from the state. The provisions of these labor agreements are binding. As a general rule, if any of the provisions are in conflict with CDCR’s policies and internal memorandums, the MOU provisions take precedence. Table 1 on the previous page presents specific language related to three labor agreements for employees who work in selected mental health and educational classifications. Two of the three MOUs prohibit CDCR (and the prison) from using “a time keeping device.” Although Mule Creek uses the accountability system as a security tool, we used the system’s data to identify the number of hours employees worked inside the prison’s secured perimeter over a specified time period; we did not use this data to support administrative actions against or pay deductions from individual employees. Timekeeping at Mule Creek State Prison Employees of Mule Creek are either paid by the hour or are salaried; most employees are paid monthly, although some are paid twice each month. Hourly employees are required to account for 40 hours of work each week, whereas salaried employees are paid a predetermined salary that is not based upon the number of hours worked but assumes that the employee will average 40 hours per week over a year. State policy requires every state employee to submit by the close of each pay period an accurate accounting of his or her actual hours worked and hours of time off; this accounting is normally referred to as a timesheet. If state employees are to be paid for their hours of time off, they must account for those hours of paid time off by using accrued personal leave. The timesheet justifies the salary and wages paid to an employee each month and is signed by both the employee and the employee’s supervisor. The signatures certify that the information is true and correct. After receiving a supervisor-approved timesheet, Mule Creek’s timekeepers review the timesheets for the accuracy of leave time used or earned and then post the leave information into a statewide system called the California Leave Accounting System (CLAS), which is maintained by the State Controller’s Office. All prison timesheets should be posted into CLAS by the tenth workday of every month in order to accurately reflect the employees’ leave hours on a monthly report generated by the prison. Using the monthly leave report, Mule Creek’s personnel specialists verify whether the timekeepers posted data from the timesheets accurately into CLAS. The personnel specialists make any necessary corrections when they find errors. Because employees are classified into many different types of job categories, monthly payroll and timesheet reconciliation is highly specialized. Most employees at Mule Creek are grouped among several different MOUs, which identify various schedules and different rules for accruing or using time off. Some employees—primarily Mule Creek’s senior managers—are excluded from collective bargaining and are not covered by an MOU. In addition, state and federal timekeeping requirements and the unique scheduling of the prison environment—24 hour days, seven days a week—result in a range of types of scheduling throughout the day. Also, generally speaking, Mule Creek’s employees work straight shifts, in which they have no official break in the middle of the shift for a meal. Bureau of Audits, Office of the Inspector General Page 8 Objectives, Scope, and Methodology In September 2010, the Office of the Inspector General received information regarding a concern with employee timekeeping and workload at Mule Creek State Prison. To ascertain the validity of this concern, we performed the following: • Reviewed laws, rules, regulations, and policies and procedures related to timekeeping and reporting. • Conducted on-site visits to observe employees at the prison. Spoke with several managers and employees. • Obtained and analyzed electronic data from several types of prison information systems, including the California Leave Accounting System, the Management Information Retrieval System, the In-Service Training Program, the Staff Accountability System, and the Case Records Automated File Tracking system. • Reviewed timesheets and work schedules for selected employees. • Reviewed Memorandums of Understanding and job duty statements for a range of employee classifications. State of California • April 2011 Page 9 Finding 1 Many of Mule Creek State Prison’s Mental Health Employees Received Full-Time Pay, But Appeared to Work Only Part Time Our analysis of several of Mule Creek State Prison’s (Mule Creek) data systems for the period of June through August 2010 revealed that many of its employees had unaccounted-for hours: that is, hours for which they were paid but which they did not spend inside the secured perimeter, in training, or in time off. We focused on certain mental health employees—the psychiatrists, psychologists, and licensed clinical social workers—because the 51 employees in these three classifications had a relatively large number of unaccounted-for hours at a very high cost: 3,386 unaccounted-for hours at a cost of $242,389 for just a three-month period. At that rate the cost would be nearly $1 million in a year. Using the prison’s data systems, we could not identify how these 51 employees spent 15 percent of their total paid hours (as shown in Figure 2). However, the data suggests that most of these employees’ unaccounted-for hours stem from arriving to Figure 2: Fifteen Percent of the Total work late and leaving early. Paid Hours for 51 Mental Health Employees for a Three-Month Period Were Unaccounted For The prison allows mental health clinicians to work four ten-hour shifts per week, and 46 of the 51 clinicians  Hours Inside the Secured have chosen this schedule. However, according to Perimeter: .................. 14,822 (66%) Mule Creek’s data for the three-month period, the 46  Hours of Authorized employees who were scheduled to work ten-hour shifts Time-off: ...................... 3,882 (17%) averaged working inside the secured perimeter only 8.4  Training Hours: ............... 374 (2%) hours per day. One of the 46 employees (a psychiatrist) averaged working inside the secured perimeter only 6.4 Unaccounted-for hours per day. The remaining five employees averaged Hours: . ......................... 3,386 (15%) working inside the secured perimeter for 7.8 hours per Total Hours Paid: ....... 22,464 day even though they were scheduled to work either eight- or nine-hour shifts. Of these five, the employee with the lowest average was a licensed clinical social worker who averaged only seven hours of an eight-hour shift. These averages are troubling, especially since we Unaccounted-for included the amount of time it took the employees to Hours Paid: walk from the front gate to their clinics after they entered 3,386 - $242,389 the secured perimeter and back again (which, generally Annualized $969,556 speaking, could take about 15 minutes each day). Some of the employees we spoke with freely admitted to leaving early, citing their Memorandum of Understanding (MOU) with the state as permission to leave early after they have completed their work. However, although Source: Mule Creek State Prison’s staff accountability, in-service training, and leave the MOUs allow for flexible work hours with prior accounting systems and payroll records supervisory approval, the MOUs do not grant permission (June through August 2010). Bureau of Audits, Office of the Inspector General Page 10 to leave early on a daily basis. In other cases, employees we spoke with provided additional partial excuses for their hours outside of the prison, explaining that they were working from home or reviewing inmate files in the administrative building. However, according to Mule Creek’s chief of mental health, working at home is not permitted. Moreover, our examination of the prison’s case records tracking system demonstrated that the mental health employees in question did not review enough inmate files in the administrative building to make much of a difference in their unaccounted-for hours. Salaried employees who leave early on a regular basis—even if they complete all of their work—essentially get paid for not working and set a poor example to their coworkers. Supervisors could instead give these employees additional assignments or responsibilities during the day to keep them fully occupied and productive, or supervisors could alter their work schedules to make their work time more efficient. Our analysis of employees’ work hours spent inside Mule Creek’s secured perimeter revealed many unaccounted-for hours Our analysis made use of Mule Creek’s electronic Staff Accountability System (accountability system), which is a card swiping system used to account for an employee’s entrance into and exit from the secured perimeter. According to the warden, the prison uses this accountability system as a security tool so that it can quickly identify which employees are located inside the secured perimeter at any given time. Although we used the system’s data to account for the number of hours certain employees spent inside the secured perimeter during the period of June through August 2010, we did not gather data to support administrative actions against or pay deductions from individual employees. We also did not gather information to measure whether individual employees completed their assignments. To draw meaningful conclusions using this data, we first determined which of Mule Creek’s employees worked inside the prison’s secured perimeter. We did this by analyzing the nature of various job assignments and their locations, interviewing prison officials, and analyzing several of the prison’s electronic data systems. The nature of certain job assignments simply did not lend themselves to our analysis because the employees did not predictably begin and end their day inside the secured perimeter or they may not have entered the secured perimeter at all. For example, employees who primarily work in the prison’s administrative building, warehouses, and minimum support facility (where minimum-security inmates are housed) would not enter the secured perimeter in any predictable manner. Further, transportation officers, relief officers, correctional counselors, building maintenance workers, and members of the warden’s and chief executive officer’s management team would legitimately—and frequently—begin or end their day outside the secured perimeter. Thus, we excluded employees in these types of positions from our analysis. Next, for the employees who we determined work primarily inside the secured perimeter, we matched names from the accountability system with Mule Creek’s electronic leave accounting, payroll, and training data. Then we calculated the amount of time the employees spent inside the secured perimeter. Since many employees are allowed to leave the secured perimeter during their shifts for various reasons, we calculated an employee’s total working hours for each day using the electronically documented time the employee first entered the State of California • April 2011 Page 11 secured perimeter and the electronically documented time the employee last exited the secured perimeter. This approach allowed the employees the most generous amount of hours each day because it included the amount of time it took for them to walk from the front gate of the secured perimeter to their assigned work area and back again (which, generally speaking, could take about 15 minutes each day). Our calculations also included the total number of hours the employees might have spent outside the prison, such as for breaks, meals, or meetings, if they entered and exited the secured perimeter several times during any particular day. To that number, we added the employees’ hours spent in training (regardless of whether the training took place inside or outside the secured perimeter), as well as the hours the prison charged employees for time off, to arrive at the employees’ total hours for the period. For some custody employees, we added the hours they were assigned to posts on a temporary basis that could take them outside of the prison on occasion. We compared the total number of hours we could account for to the number of regular, overtime, and excess hours for which the employees were actually paid during the same period. Unaccounted-for hours represent the difference between the number of hours we could account for and the number of hours the employee was paid during the three-month period. Using this approach, we identified all or nearly all of the hours for 692 of the 827 employees who met our criteria. However, the remaining 135 employees, who worked in a variety of classifications, had a relatively high number of unaccounted-for hours. After this initial analysis, we focused on certain mental health and educational employees, specifically the Some explanations for employees having psychiatrists, psychologists, licensed clinical social unaccounted-for hours: workers, academic teachers, vocational instructors, • Arriving to work late and/or leaving early. and educational supervisors, because these (Collectively, this is the single largest component of employees’ unaccounted-for employee groups stood out as having relatively hours.) large numbers of employees with unaccounted-for • Failing to report full days of time-off on a hours at a relatively high financial cost.1 The text timesheet. (This timekeeping mistake is box at left identifies some explanations for these discussed in Finding 3.) employees having unaccounted-for hours. After • Recording the incorrect number of time-off factoring the number of hours these employees hours in the California Leave Accounting System. (This timekeeping mistake is were short each day and the number of days each discussed in Finding 3.) of them worked inside the secured perimeter, our • Working on tasks outside of the secured analysis suggests that the single largest cause of perimeter. (This is not common since the unaccounted-for hours stemmed from employees employees we evaluated worked primarily not averaging a full day of work inside the secured inside the secured perimeter.) perimeter. However, combinations of other • Forgetting an identification card for a day. (This rare occurrence explains not factors may also have resulted in unaccounted- having work hours recorded in the Staff for hours. For example, some employees also had Accountability System.) unaccounted-for hours from data entry mistakes into the prison’s leave accounting system or from 1We discuss academic teachers, vocational instructors, and educational supervisors in Finding 2 of this report, beginning on page 23. Bureau of Audits, Office of the Inspector General Page 12 unpredictably working outside the prison’s secured perimeter (where the accountability system would not track the employees’ hours). Nearly all of Mule Creek’s psychiatrists regularly arrived to work late and left early Mule Creek psychiatrists are salaried employees and exempt from receiving overtime pay. Figure 3 on the following page shows each of the psychiatrist’s hours inside the secured perimeter on a typical workday during the period of June though August 2010. Eleven psychiatrists were scheduled to work ten-hour shifts and two were scheduled to work eight- hour shifts. All but one of the psychiatrists’ ten-hour shifts began by 8:00 a.m., yet only one psychiatrist typically entered the prison by this time. Similarly, although all but one of the psychiatrists’ shifts ended at 6:00 p.m., most of the psychiatrists regularly left the prison well before 4:30 p.m. Consequently, 11 psychiatrists who were scheduled to work ten-hour shifts spent, on average, only 7.5 hours inside the secured perimeter each day (two and a half hours short of a full shift). These employees’ time spent inside the secured perimeter ranged between 6.4 and 8.4 hours per day. Although these 11 psychiatrist’s daily averages equate to their working approximately 26 to 34 hours of their 40 full-time hours per week, they were still paid for working full time.2 In contrast, the two psychiatrists who were scheduled to work eight- hour shifts averaged working 40 hours per week. Psychiatrists are among the highest-paid employees in the prison, with salaries averaging $245,000 per year. Consequently, Mule Creek must ensure these employees deliver the most effective services to inmates who require their services as well as administer the most effective use of the state’s resources. For the period of June through August 2010, Mule Creek scheduled psychiatrists to begin seeing their first patients by 8:30 a.m. and to complete seeing their last patients by 4:00 p.m. Psychiatrists typically had a break from seeing patients between 11:30 a.m. to 12:30 p.m. each day, while the prison conducted an inmate count. Generally speaking, this schedule resulted in only six and a half hours each day of patient appointment time and three and a half hours each day of non-patient or administrative time. According to Mule Creek’s chief psychiatrist, the prison’s psychiatrists should be using the time between 4:00 p.m. and 6:00 p.m. to review charts, perform evaluations, dictate notes, and complete other work assignments. However, in some cases, our discussions with psychiatrists revealed something else entirely. To understand the psychiatrists’ work schedules and responsibilities and to ask them about their work hours during this three-month period, we spoke with six of the 11 psychiatrists who appeared to work substantially fewer hours than they were scheduled to work. We gave each of the employees an opportunity to provide a response to our queries, but we did not compel any of them to speak with us. Three of the six psychiatrists freely admitted to leaving early and confidently asserted that their MOU allowed them to leave for the day after they had completed their work. Although the MOU states (in pertinent part) that these employees have flexibility to alter their daily and weekly work schedules, this flexibility does not allow them to leave early 2One of the 11 psychiatrists was scheduled to work ten-hour shifts only two days per week (half time). This person was still paid a salary during our three-month time period, but did not work all of his or her scheduled hours inside of the secured perimeter during the period. For comparative purposes, we adjusted this person’s average workweek to a full-time schedule. State of California • April 2011 Page 13 Figure 3: Average Daily Hours Mule Creek’s Psychiatrists Worked Inside the Secured Perimeter (June through August 2010) 8:30 A.M. 4:30 P.M. 8 A.M. 4 P.M. 6 P.M. Ten-hour Shifts Psychiatrist 1 6.4 hr./day - 25.6 hr./wk. Psychiatrist 2 7.1 hr./day – 28.4 hr./wk. Psychiatrist 3 7.1 hr./day – 28.4 hr./wk. Psychiatrist 4 7.2 hr./day – 28.8 hr./wk. Psychiatrist 5◊ 7.6 hr./day – 30.4 hr./wk. Psychiatrist 6 7.6 hr./day – 30.4 hr./wk. Psychiatrist 7 7.7 hr./day – 30.8 hr./wk. Psychiatrist 8 7.8 hr./day – 31.2 hr./wk. Psychiatrist 9§ 7.9 hr./day – 31.6 hr./wk. ◊ This psychiatrist was scheduled to work ten-hour shifts for only two days per week (half time). For comparative purposes, we adjusted his or her Psychiatrist 10 8.1 hr./day – 32.4 hr./wk. average workweek to a full-time schedule. Psychiatrist 11 8.4 hr./day – 33.6 hr./wk. § These psychiatrists are supervisors. ‡ The averages are weighted based on the number Average‡ 7.5 hr./day – 30.0 hr./wk. of days each employee worked. † We calculated an employee’s average daily work Eight-hour Shifts hours on typical work days by taking the number Psychiatrist 12§ 7.9 hr./day – 39.5 hr./wk. of hours between the average time of day the employee first entered the secured perimeter and Psychiatrist 13 8.3 hr./day – 41.5 hr./wk. the average time of day the employee last exited the secured perimeter. We calculated the average workweek by multiplying the average daily work Average‡ 8.0 hr./day – 40.0 hr./wk. hours by the number of regularly scheduled workdays in a week. The actual number of hours scheduled work hours Start/End Time — worked might vary due to the amount of training or average daily work hours† time off used by the employee. Source: Mule Creek State Prison’s staff accountability system for the period of June through August 2010, and employee work schedules. on a regular basis without working longer hours at another time to balance their hours. For example, the agreement establishes an expected 40-hour workweek in the following statement: While full time [psychiatrists] will be scheduled to work an average of forty (40) hours per week during 12 consecutive pay periods, the actual daily and weekly work schedule may vary due to time requirements of specific assignments. Inherent in their job is the responsibility and expectation that work weeks of longer duration may be necessary. (7.6(A)) (Emphasis added) Nonetheless, as Figure 4 on the following page shows, 11 of the psychiatrists collectively worked inside the secured perimeter a full day of at least ten hours only 4 percent of their work days in the three-month period ending August 2010. In addition, the psychiatrists who were scheduled to work 40 hours per week in four ten-hour shifts did not work inside the secured perimeter for Bureau of Audits, Office of the Inspector General Page 14 that amount of time, but instead worked substantially less, averaging as a group roughly 30 hours per week. One psychiatrist even commented, “I don’t work a full ten-hour day, but I get paid for ten hours.” Although their MOU allows longer weeks to make up for shorter weeks, we find it improbable that the psychiatrists would regularly work more than their scheduled hours per week just to overcome their deficit hours from earlier weeks, especially in light of the limited number of daily patient Figure 4: Eleven Psychiatrists hours available to them. Collectively Worked a Full Day of Ten Hours Only 4 Percent of the Time During our discussion, one of the psychiatrists accepted Full Working Days of responsibility for being short of hours and asked that the At Least Ten Hours: difference in hours be deducted from his or her annual 4% (14 days) leave balance. Another psychiatrist politely argued that the MOU includes on-call hours as part of his or her 40-hour workweek; however, based on our review of the agreement, we find that this reading does not appear accurate. The MOU provides that an employee’s At Least average workweek “shall not include any hours of Eight Hours work performed while on Medical Officer of the Day But Less Than assignments…or On-Call/Call Back Assignment.” Ten Hours: 33% (115 days) Moreover, two psychiatrists explained that, on occasion, Fewer Than they go to Mule Creek’s administration building to Eight Hours: review inmate files for their cases, which may account 63% for a small portion of their unaccounted-for hours. (224 days) However, one of those employees implied that his or her review of inmate files would typically take place during the middle of his shift—which we would have Source: Mule Creek State Prison’s staff accountability and in-service training systems already counted—rather than at the end of his shift. (June through August 2010). To understand the frequency of clinicians’ performing these types of file reviews in the administrative building, we spoke with some of the case records employees in that building. These employees check inmate files out to clinicians. Generally speaking, the case records employees told us that mental health employees do not frequently review files. Two case records employees estimated that individual mental health clinicians may review a file once or twice a month. To corroborate this, we obtained the case records office’s electronic data, which is used to track inmate central files checked out to employees of the prison. Our analysis of this data for the period of June through August 2010 found that none of the files were specifically checked out by name to any of the psychiatrists, psychologists, or licensed clinical social workers. Some files were checked out anonymously, using a code that identifies a file review that takes place at the counter or at a desk in the case records office by any employee. On average, fewer than four anonymous on-site file reviews took place in the case records office each day. Even assuming that all anonymous file reviews during this period were only made by the 51 mental health employees, we find that at the rate of fewer than four reviews per day, these reviews would not have accounted for a significant part of a mental health clinician’s day. Further, since the case records office closes at 4:30 p.m., the clinicians could not have reviewed inmate files in the case records office between the hours of 4:30 p.m. and 6 p.m. State of California • April 2011 Page 15 Three psychiatrists claimed to work from home occasionally, performing tasks such as dictating and researching. However, when we spoke to Mule Creek’s chief of mental health about this, he indicated that working at home is not permitted, adding, “There is no need that I am aware of” for psychiatrists and psychologists to work at home. The chief of mental health said he would be concerned about inmate-patient confidentiality if the clinicians did in fact work from home. One of the three psychiatrists gave a few examples of outside tasks he performs but then told us, “I do enough outside activities to make up a 40-hour workweek, but I will not specify those activities to you.” However, given the chief of mental health’s position against working at home, we are unsure of what those activities could be, and we are therefore suspicious of the psychiatrist’s claim. Managers knew that their subordinates were leaving early, but described difficulties holding them accountable We asked the six psychiatrists if their supervisors were aware of their arriving to work late and leaving work before completing their scheduled work hours. One clinician commented that it is just “part of the culture” at Mule Creek to leave after patient work is completed. A clinician with a long commute of approximately three hours explained that his or her supervisor is aware of his or her special circumstances, and therefore maintains a permissive attitude toward lateness. This person also stated that there is a feeling among staff that it is permissible for them to leave at 4:15 p.m. because coverage from the on-call shift for psychiatrists begins at 4:00 p.m. In one case where the psychiatrist’s supervisor was present for a portion of the discussion, the supervisor acknowledged that some employees take advantage of the system and do not work their full schedule. When we brought this matter to the attention of the prison’s chief executive officer, the chief of mental health services, and the chief psychiatrist, all three of them expressed concern about their employees’ not working full-time hours inside the prison. Yet each of them also acknowledged that they were not surprised by our findings. The chief executive officer said that he was aware that mental health employees in particular have not been held accountable. In addition, the chief executive officer indicated that not only is there a culture of leaving early among mental health employees at Mule Creek, but the issue is pervasive at another prison where he is also the chief executive officer. We are therefore concerned that the problems we found at Mule Creek could potentially exist at another prison, or even worse, at prisons statewide. The chief of mental health services explained that he has had difficulties holding employees accountable for their work hours and added that, short of standing at the gate all the time, “There is nothing I can do about it.” Shortly after we began our fieldwork, he wrote a letter in October 2010 to mental health staff revising the start times of employees working alternative work schedules (such as four ten-hour shifts each week totaling 40 hours worked, or 4/10/40) and acknowledged that “working hours in the past have been very relaxed, but at this time, we need to adjust to the increasing needs of [mental health] programs.” Ironically, the chief of mental health himself appears to work fewer than his own scheduled hours inside the secured perimeter. According to our analysis, he only worked an average of about 7.6 hours of his scheduled nine- hour shifts inside the prison. He defended his apparently short hours by explaining that he visited the administration building often at the end of his day and occasionally watched the front gate to see if employees were leaving early. When we inquired further about this claim, we learned that Bureau of Audits, Office of the Inspector General Page 16 the chief did not keep any records of dates when he watched at the front gate or of employees whom he witnessed leaving early; the lack of documentation, of course, severely limits the potential effectiveness of his actions. Nevertheless, if the claim is true, we question the need for the chief of mental health to spend 70 hours (his number of unaccounted-for hours) over a three- month period in the administrative building or at the front gate when his staff and office space are located inside the secured perimeter. The chief psychiatrist also acknowledged having trouble holding employees accountable. He told us that he has attempted on several occasions to clarify which remedies, if any, are available to managers to prevent psychiatrists from leaving Mule Creek at their own discretion. He was told that he could not simply order a psychiatrist to return to the prison on his or her regular day off to “make up” time lost through leaving early on a previous day. In other words, if a psychiatrist shows up for any portion of a shift, he or she is credited for having worked the whole shift. He further described his difficulties in attempting to discipline an employee who only worked from 8:30 a.m. to 12:00 p.m. on one day, but claimed to have seen ten patients lasting five hours, which, as the chief pointed out, is not possible in that time period. He said he was told that he did not have sufficient grounds to pursue adverse action against the employee, which could have resulted in monetary penalties. Instead, he was told he could issue a letter of instruction to the employee, which effectively serves as a warning. The chief psychiatrist concluded that the current progressive discipline process at Mule Creek is not well-suited to correcting psychiatrists’ performance, productivity, or accountability. The chief psychiatrist claimed that the progressive discipline process is not well-suited for addressing employee performance issues. However, we note that, by identifying a behavior that was not desirable and then counseling the employee, the chief took the first steps of the progressive discipline process. As a second step, he should continue to monitor the psychiatrist’s workload and length of time of the clinician’s appointments and immediately correct any unacceptable behaviors when observed. Although it is true that the chief cannot require a psychiatrist to come in on a regularly scheduled day off to “make up” time lost through leaving early on another day, he can require the employee to stay late or come in on his or her day off to complete assignments that were unfinished or neglected due to leaving early. The chief can also assign his employees appropriate tasks that require them to work an average of 40 hours per week. Consistent with the employees’ MOU, the chief psychiatrist can establish the psychiatrists’ work hours and the expectation that arriving on time is essential and leaving early is not permitted without prior supervisory approval. The MOU presents this provision in the following language: Management can require [psychiatrists] to work specified hours. Subject to prior notification and management concurrence, [psychiatrists] have the flexibility to alter their daily and weekly work schedules. [Psychiatrists] are responsible for keeping management apprised of their schedule and must receive prior approval from management for the use of accrued leave for absences of any duration. (7.6(C)(4)) There are other proactive actions the chief psychiatrist can perform to assist his employees in becoming more efficient and productive. He can monitor employee time by physically checking in with the employees in the morning and before the end of the day and record any instances of noncompliance with established work hours. Ultimately, these actions would assist State of California • April 2011 Page 17 the chief psychiatrist in holding his employees more accountable for their work hours while maximizing patient hours. Psychologists and licensed clinical social workers also worked fewer hours inside the secured perimeter than they were paid to work As was true among the psychiatrists, a high percentage of Mule Creek’s psychologists and licensed clinical social workers also worked fewer hours inside the prison than the number of hours for which they were paid. Figure 5 shows the collective average number of hours these two classifications spent inside the secured perimeter by duration of shifts for the three- month period, ending August 2010. These employees were scheduled to work 40 hours per week during this period among up to three different shift durations, yet as groups, all but one (having only one employee) averaged 35 hours or less per week. This includes 26 of the 31 psychologists who individually averaged working inside the secured perimeter the equivalent of only 28 to 39 hours per week. Further, all seven of the licensed clinical social workers averaged less than a full day’s work inside the secured perimeter, collectively averaging the equivalent of 33 hours per week. The range of these employees’ hours inside the secured perimeter was the equivalent of between 28 and 38 hours per week. During our review, we attempted to speak with eight psychologists and six licensed clinical social workers. We gave each of the employees an opportunity to provide a response to our queries, but we did not compel any of them to speak with us. In fact, two of the psychologists refused to answer our questions. One psychologist refused to speak with us, stating that he did not want to answer questions without his attorney present because he did not want to incriminate himself. The other psychologist who refused to speak with us offered a similar response. One psychologist and one licensed clinical social worker with whom we spoke explained that, on occasion, they go to the administration building to review inmate central files for cases; however, Figure 5: Comparison of Psychologists’ and Licensed Clinical Social Workers’ Average Daily Work Hours Inside the Secured Perimeter (June through August 2010) Psychologists 8 9 10 hour day † This employee was scheduled to work eight bbbbbbbbbbbbbbbbbbbb nine-hour shifts and one eight-hour shift every bbbbbbbbb 8.7 hr./day - 34.8 hr./wk. two weeks (9/8/80). b ‡ We calculated a classification’s average daily 7.6 hr./day - 34.2 hr./wk.† work hours on typical work days by taking b the number of hours between the average 8.6 hr./day - 43.0 hr./wk. time of day the employees first entered the Licensed Clinical Social Workers secured perimeter and the average time of day the employees last exited the secured bbbbbb perimeter. The averages are weighted based 8.2 hr./day - 32.8 hr./wk. on the number of days the employees in each b classification worked. 7.0 hr./day - 35.0 hr./wk. We calculated the average workweek by b = Number of employees scheduled work hours multiplying the average daily work hours by the number of regularly scheduled workdays average daily work hours‡ in a week. The actual number of hours worked might vary due to the amount of training or time off used by the employees. Source: Mule Creek State Prison’s staff accountability system for the period of June through August 2010, and employee work schedules. Bureau of Audits, Office of the Inspector General Page 18 as previously discussed, based on the low frequency of file reviews, we believe that this would account for only a small portion of their time at best. One of these clinicians indicated that their MOU allowed clinicians to leave early after they complete their work; again, however, because these clinicians’ MOU is similar to the psychiatrists’ MOU, it does not appear to provide such leniency. Overall, five clinicians told us that their supervisor is aware that they leave early. By switching to an eight-hour workday, Mule Creek could increase mental health patient hours by up to 25 percent Most of Mule Creek’s psychiatrists, psychologists, and licensed clinical social workers work alternative work schedules, which consists mostly of a four-day-per-week schedule of ten-hour shifts (4/10/40) and, in one case, eight days of nine-hour shifts over two weeks along with one eight-hour day (9/8/80). Beginning on November 29, 2010, clinicians began scheduling patients’ appointments Monday through Friday from 8:00 a.m. at the earliest and completing their last patient appointment by 4:00 p.m. (an increase of 30 minutes per day). Clinicians still typically take a break from their appointments for approximately one hour from 11:30 a.m. to 12:30 p.m. each day while the prison conducts an inmate security count. Appointments can also be canceled, moved, or rescheduled for various reasons. Therefore, clinicians who are scheduled to work four ten-hour shifts are currently able to see inmates for only about seven hours per day. This leaves these clinicians with approximately three hours per day of non-patient time, or administrative time. We recognize that clinicians are supposed to work on file reviews, inmate evaluations, dictations, and other tasks during their administrative time, but we found that this is not always the case. As we discussed previously, some clinicians we spoke with told us that they often leave work after they complete their appointments with patients instead of remaining at the prison until their scheduled shift ends. These comments are consistent with our data analysis, which also shows clinicians working fewer hours inside the secured perimeter than they were paid to work. Therefore, since clinicians who work a four ten-hour shift cannot see patients for a sizable portion of their day, we question whether clinicians’ having a ten-hour work schedule provides the most efficient arrangement for maximizing patient care. Our analysis suggests that clinicians at Mule Creek could meet more frequently with patients by working a five-day-per-week, eight-hour schedule instead of the current four-day, ten-hour schedule almost all of them preferred. Using a simple illustration, we demonstrate in Figure 6 on the following page how this change could result in an increase of up to a 25 percent in available patient hours. For example, in a four-day workweek (for illustration purposes, we chose Monday through Thursday), each clinician under his or her current schedule can only meet with patients for 28 of the 40 weekly hours, or 70 percent of the time. Clinicians are then left with three hours each day of non-patient time, or administrative time, for a total of 12 hours per week, or 30 percent of their weekly scheduled work hours. However, if the clinicians switched to an eight- hour shift five days per week, the more traditional schedule would yield seven hours of additional patient time per week per employee. According to the prison’s records for June through August 2010, a total of 46 clinicians—29 psychologists, 11 psychiatrists, and 6 licensed clinical social workers—worked ten-hour shifts.3 3Although one psychiatrist was scheduled to work a ten-hour shift, two days per week, we have included that individual in our analysis as working full time for illustration purposes only. State of California • April 2011 Page 19 Figure 6: Mental Health Clinicians Could Increase Their Available Patient Hours by 25 Percent by Switching to an Eight-Hour Day Current schedule: Proposed schedule: Four-day, ten-hour schedule (4/10/40) Five-day, eight-hour schedule (5/8/40) 10 9 8 7 6 5 4 3 2 1 M T W Th F M T W Th F Benefits of changing to 5/8/40 schedule: Non-patient hours • Percentage increase in patient hours from 4/10/40 .................................................... 25% • Percentage reduction of non-patient hours Patient hours from 4/10/40 .................................................... 58% 4/10/40 schedule: 5/8/40 schedule: Total weekly patient hours .................................... 28 Total weekly patient hours .................................... 35 Total weekly non-patient hours ............................. 12 Total weekly non-patient hours ............................... 5 Total hours per week ............................................ 40 Total hours per week ............................................ 40 Source: Office of the Inspector General’s analysis of Mule Creek State Prison’s work schedules. As shown in Table 2 on the following page, if all of the 46 clinicians at Mule Creek who worked a four-day, ten-hour shift changed their schedules to work five-day, eight-hour shifts, this change would yield approximately 322 more patient hours per week (or 644 more 30-minute appointments). It would also yield 1,288 more patient hours per month (or 2,576 more 30-minute appointments). Furthermore, with only one hour per day of unscheduled administrative time, the total number of non-patient hours per week would decrease from 12 hours to only five hours (a decrease of 58 percent) per clinician. The concept of changing the clinicians’ schedule was even brought up by one of the psychiatrists we spoke with, who noted that the four-day, ten-hour shift is not productive for patient care. He said, “There is no reason for any of the psychiatrists to be on a 4/10 work schedule. Unless there is charting to finish, there is nothing productive to do from 4 p.m. to 6 p.m.” The psychiatrist’s comment appears particularly accurate since our data analysis reveals that many of the mental health employees leave Mule Creek early on a regular basis. In fact, the 46 clinicians who were scheduled to work ten-hour shifts averaged working inside the secured perimeter as a group for only 8.4 hours per day. Since these employees are salaried, we recognize that they have the Bureau of Audits, Office of the Inspector General Page 20 Table 2: Summary of Additional Patient Hours and Appointments With an Extra Seven Hours per Clinician Employee Total Additional Patient Hours Classification Employees per Week per Month Psychologists 29 203 812 Psychiatrists 11 77 308 Licensed Clinical Social Workers 6 42 168 Total 46 322 1,288 Additional Patient Appointments per Month (30 minutes per visit) 644 2,576 Source: Office of the Inspector General’s analysis of Mule Creek State Prison’s personnel records for the period of June through August 2010. flexibility to alter their daily work schedules based on their workload. However, we believe that a change in their schedule would help maximize their time with patients, reduce administrative time and perhaps diminish their reasons for leaving early. The five clinicians who were scheduled to work eight- or nine-hour shifts help illustrate this point because, as a group, they averaged working a higher percentage of their shifts inside the secured perimeter. These five clinicians averaged 7.8 hours per day inside the secured perimeter. When we discussed the idea of changing the clinicians’ schedules with the chief executive officer, the chief of mental health, and the chief psychiatrist at the prison, all three were generally supportive. All three agreed that more productive hours would be achieved in five eight-hour days each week instead of four ten-hour days. The chief executive officer even said, “The 4/10 schedule needs to go away.” Although the chiefs of mental health and psychiatry also supported the idea, they expressed some concern that changing work schedules would not be well received by some employees, and they warned that mental health services at Mule Creek could lose some employees if the prison implemented a change to employees’ work schedules. Nevertheless, we believe that such a change is necessary to better align clinicians’ work schedules with delivering optimal patient care. Eight-hour shifts offer more potential to maximize clinical hours while minimizing administrative time, and potentially, the reasons for employees to leave work before their shift ends. Recommendations: To ensure that Mule Creek State Prison receives the most productivity and value from its psychiatrists, psychologists, and licensed clinical social workers, the Office of the Inspector General recommends that the prison’s chief executive officer, in accordance with and to the extent permissible under his employees’ labor agreements, take the following actions: • Develop a method to hold supervisors and managers in mental health services accountable for ensuring that their employees complete required daily tasks and obtain prior permission before the employees leave work early. • Require all supervisors to maintain logs tracking the number of times employees request to leave early and, when necessary, adjust employees’ workload accordingly. State of California • April 2011 Page 21 • Investigate and take administrative actions against employees who leave work before completing their daily tasks or without prior permission from their supervisors. • Train all supervisors and managers on how to initiate progressive discipline on employees when appropriate. To ensure that the state receives the most productivity and value from its mental health clinicians, the Office of the Inspector General recommends that the California Department of Corrections and Rehabilitation, Division of Correctional Health Care Service (DCHCS), in accordance with the employees’ labor agreements, take the following action: • Evaluate the effectiveness of delivering mental health services to inmates using alternative work schedules for mental health clinicians at all state prisons and, in cases where DCHCS determines that alternative work schedules deliver fewer patient services or are otherwise less effective than a traditional five-day-per-week schedule, eliminate their usage. Bureau of Audits, Office of the Inspector General Page 22 Finding 2 Mule Creek State Prison’s Academic Teachers, Vocational Instructors, and Educational Supervisors Appeared to Work Less Than Full Days Our analysis of Mule Creek State Prison’s (Mule Creek) Staff Accountability System (accountability system) for the three months ending August 2010 revealed that 12 full-time academic teachers and five full-time vocational instructors averaged working less than a full eight-hour shift inside the prison’s secured perimeter. For example, the academic teacher with the lowest average worked inside the secured perimeter for only 6.6 hours per day (or the equivalent of 33 hours of his or her scheduled 40 hours per week). Also troubling is that their educational supervisors—including the principal and the two vice principals—also put in less than a full day’s work, averaging between 6.6 and 7.0 hours per day of their eight-hour shifts (or the equivalent of 33 to 35 hours of their scheduled 40 hours per week) inside the secured perimeter. Since all of these employees were paid for working full time, the total cost of their 793 unaccounted-for hours is Figure 7: Eight Percent of the $30,511 for the three-month period, or if this rate were Total Paid Hours for 20 Education constant for 12 months, $122,044 for the year (refer to Employees for a Three-Month Period Figure 7). During our conversations with several of the Were Unaccounted For academic teachers and vocational instructors about the  Hours Inside the Secured number of hours they typically worked, some claimed Perimeter: .................... 6,743 (63%) that they worked outside the prison for about one hour  Hours of Authorized each day; they called this “off-site preparation time.” Time-off: ...................... 2,984 (28%) One of the employees claimed that he used his off-site  preparation time as part of his daily commute. According Training Hours: ................. 40 (<1%) to the principal, off-site preparation time must be formally Unaccounted-for requested and approved by the vice principals in advance. Hours: ............................. 793 (8%) However, our conversations with the vice principals Total Hours Paid: ....... 10,560 revealed that teachers and instructors did not frequently request to work off site. Consequently, Mule Creek must improve its management of these employees’ work hours. Academic teachers and vocational instructors claimed to Unaccounted-for work off-site, but it appears they did so without permission Hours Paid: We analyzed the average time of day that academic 793 - $30,511 teachers, vocational instructors, and educational Annualized: $122,044 supervisors entered and exited Mule Creek’s secured perimeter for the three months ending August 2010. We spoke to four academic teachers and two vocational instructors as well as the principal and two vice principals at the prison. We learned that all teachers and instructors are scheduled to work an eight-hour shift between the Source: Mule Creek State Prison’s staff accountability, in-service training, and leave hours of 6:45 a.m. and 2:45 p.m. Within this period, accounting systems and payroll records teachers have instructional time with inmates between the (June through August 2010). State of California • April 2011 Page 23 hours of 7:45 a.m. and 2:15 p.m., including a 30- to 45-minute break between morning sessions and afternoon sessions. This leaves teachers with about two hours each day to complete non- instructional activities, including preparation tasks that should take place inside the prison. Such tasks could include planning lessons, grading assignments, or conducting research. Using data from Mule Creek’s accountability system, Figure 8 displays the average number of hours that the 12 academic teachers and five vocational instructors worked between June and August 2010. Using the average time of day that the employees’ first entered and last exited the Figure 8: Average Daily Hours Mule Creek’s Educators Worked Inside the Secured Perimeter (June through August 2010) 7:45 A.M. 2:15 P.M. 6:45 A.M. 2:45 P.M. Academic Teachers Teacher 1 6.6 hr./day – 33.0 hr./wk. Teacher 2 6.7 hr./day – 33.5 hr./wk. Teacher 3 7.1 hr./day – 35.5 hr./wk. Teacher 4 7.2 hr./day – 36.0 hr./wk. Teacher 5 7.3 hr./day – 36.5 hr./wk. Teacher 6 7.3 hr./day – 36.5 hr./wk. Teacher 7 7.6 hr./day – 38.0 hr./wk. Teacher 8 7.6 hr./day – 38.0 hr./wk. Teacher 9 7.7 hr./day – 38.5 hr./wk. Teacher 10 7.8 hr./day – 39.0 hr./wk. Teacher 11 7.8 hr./day – 39.0 hr./wk. Teacher 12 7.8 hr./day – 39.0 hr./wk. Average‡ 7.4 hr./day – 37.0 hr./wk. Vocational Instructors Instructor 1 6.8 hr./day – 34.0 hr./wk. Instructor 2 ‡ The averages are weighted based on the number of 6.9 hr./day – 34.5 hr./wk. days each employee worked. Instructor 3 7.2 hr./day – 36.0 hr./wk. † We calculated an employee’s average daily work hours on typical work days by taking the number Instructor 4 7.4 hr./day – 37.0 hr./wk. of hours between the average time of day the employee first entered the secured perimeter Instructor 5 7.5 hr./day – 37.5 hr./wk. and the average time of day the employee last exited the secured perimeter. We calculated the Average‡ 7.2 hr./day – 36.0 hr./wk. average workweek by multiplying the average daily work hours by the number of regularly scheduled workdays in a week. The actual number of hours scheduled work hours Start/End Time — worked might vary due to the amount of training or average daily work hours† time-off used by the employee. Source: Mule Creek State Prison’s staff accountability system for the period of June through August 2010, and employee work schedules. Bureau of Audits, Office of the Inspector General Page 24 secured perimeter, we note that these employees typically averaged about 6.6 to 7.8 hours per day inside the secured perimeter, which is the equivalent of about 33 to 39 hours per week. We also note that, during the core hours of instruction—between 7:45 a.m. and 2:15 p.m.—only seven of the 12 academic teachers and one of the five vocational instructors were typically present inside the secured perimeter for the full duration. The other nine employees typically exited the secured perimeter before classes were scheduled to end. To understand the reasons for the apparent shortage in daily work hours, we spoke with four academic teachers, two vocational instructors, the two vice principals, and the principal at Mule Creek. We gave each of the employees an opportunity to provide a response to our queries, but we did not compel any of them to speak with us. The four academic teachers we spoke with told us that they work one hour outside of the prison each day while they conduct their preparation activities. Two of these teachers said that they used their hours at home to conduct research because there is a lack of resources at the prison, including internet access. However, neither teacher could estimate how frequently they used the internet for research nor did they keep records of their activity. Without documentation of their activities, the prison cannot assess the usefulness of this time nor hold the teachers accountable if their supervisors later determine the time was wasted. Another teacher commented that he uses his commute time as part of his preparation time. He said, “I travel to work during my commute and think about work. The idea is I think about work during my commute.” This employee further explained that his supervisor is aware of his travel preparation time and even allows the practice. However, when we asked the employee’s supervisor about this, she emphatically denied having approved this employee’s off-site time. Another academic teacher explained that he teaches at the minimum support facility for about two to three hours per week, which could account for a portion of his shortage of hours since the accountability system would not account for his hours at that location. The educators’ Memorandum of Understanding (MOU) allows for off-site preparation time if management approves of the activity. The MOU also clarifies the use of preparation time in the following provision: During a teacher’s workday, there shall be scheduled non-instructional periods for purposes of teacher preparation and for performance of other job duties. Teacher preparation is work time to be used for the purpose of supporting classroom instruction at a level consistent with the diversity of student needs and changing program demands. (21.14(A)) When we asked the principal about academic teachers and vocational instructors leaving work early for preparation activities, he responded that the implied statewide standard for preparation time at all prisons is one hour per day, and added that he was not aware of any employee using off-site preparation time on a regular basis. When asked if academic teachers and vocational instructors could use their preparation time off-site, he answered, “No. The one hour of preparation time has to be completed on institutional grounds.” He offered the following explanation: There is no need for academic teachers to use off-site preparation time, everything needed is on grounds. For vocational teachers, some off-site preparation time may State of California • April 2011 Page 25 be needed for their Trade Advisory Council (TAC)4 activities in the community, but would not be needed regularly. Regardless if the teacher is an academic or vocational teacher, any off-site preparation time must be formally requested by the teacher to their vice principal, and the vice principal must approve the off-site preparation time. The vice principal of the vocational instructors said he was aware that both academic teachers and vocational instructors leave early. In addition, he only remembered receiving about three to four off-site preparation requests for TAC meetings during our review period. The vice principal indicated that he thought academic teachers and vocational instructors leave early because they are considered “exempt” employees, not because they are conducting off-site preparation activities. However, he added that he thought academic teachers and vocational instructors only leave when their work and student contact is complete. He expressed frustration over the difficulty he experienced in holding employees accountable, saying that he “would try to make them stay,” but noted that “once the students are released and all required documentation is completed, there is no defined reason to require staff to remain on-site.” Similarly, the vice principal of the academic teachers commented that she, too, is unhappy with academic teachers and vocational instructors leaving early and claiming it as off-site preparation time. Then she added, “There is nothing I can do about it.” Despite both vice principals’ lack of confidence in their ability to hold their employees accountable for working a full shift, the vice principals could establish additional activities for their employees that would keep them productive rather than allow them to leave early. As the principal said, “There is plenty of work to keep staff busy.” And as explained in the previous section, managers can take steps to hold their employees accountable if the employees are not completing assignments. As long as there are tasks to complete, managers can require employees to work an average of 40 hours per week. The principal and the two vice principals also worked fewer than eight hours per day Not only does our data analysis show that academic teachers and vocational instructors at Mule Creek left the secured perimeter early, but it also reveals that their educational supervisors—the principal and the two vice principals—left early as well. Figure 9 on the following page shows that during the three-month period under our review, the principal and the two vice principals averaged only 6.6 to 7.0 hours inside the secured perimeter during their eight-hour shifts, or the equivalent of 33 to 35 hours per week. We spoke with the principal and the two vice principals to understand their perspective of our data analysis. The principal explained that he frequently visited the administration building in the morning, sometimes for as long as 20 minutes, but also admitted that he left work early on Mondays when teachers were furloughed and on Fridays a couple of times each month to travel out of town on personal business. The vice principal of vocational instructors indicated that he checked his mail 4The TAC is intended to provide positive benefits to CDCR in the area of trade and industrial training programs. The warden, where possible, establishes the TACs for vocational training programs in the area of vocational education, industries, maintenance, or food service. The TAC provides standards for inmate selection and training that are relevant to current methods and industry standards. Bureau of Audits, Office of the Inspector General Page 26 Figure 9: Average Daily Hours Mule Creek’s Education Supervisors Worked Inside the Secured Perimeter (June through August 2010) 7:30 A.M. 3:30 P.M. 7 A.M. 3 P.M. Principal and Vice Principals † We calculated an employee’s average daily work hours Supervisor 1 6.6 hr./day – 33.0 hr./wk. on typical work days by taking the number of hours between the average time of day the employee first Supervisor 2 6.9 hr./day – 34.5 hr./wk. entered the secured perimeter and the average time of day the employee last exited the secured perimeter. Supervisor 3 7.0 hr./day – 35.0 hr./wk. We calculated the average workweek by multiplying the average daily work hours by the number of regularly scheduled workdays in a week. The actual number of scheduled work hours Start/End Time — hours worked might vary due to the amount of training average daily work hours† or time off used by the employee. Source: Mule Creek State Prison’s staff accountability system for the period of June through August 2010, and employee work schedules. at the administration building and would sometimes exit the secured perimeter early for meetings in the administrative building, to handle inmate appeals, or to work in the prison’s minimum security facility, which is outside the secured perimeter. The vice principal of academics indicated that she did not know how to explain her unaccounted-for hours, and indicated that she attended some medical appointments and ran errands around the prison’s administrative building and warehouses. Both vice principals noted that they are salaried employees and that their MOU with the state allows them to leave the prison after they complete their work. Their MOU allows them to alter their work hours (with prior management approval), and it also states (in part) that they “are expected to work, within reason, as many hours as necessary to accomplish their assignments or fulfill their responsibilities…” However, the MOU clearly points out that their workweek shall be 40 hours. Therefore, this language does not suggest that these employees can simply leave early on a regular basis. Nevertheless, when the principal and vice principals regularly leave early, their actions set a poor example for employees to follow. Furthermore, we question these supervisors’ ability to hold employees accountable for working a full shift when they do not even hold themselves to the same standard. Recommendations: To ensure that Mule Creek State Prison receives the most productivity and value from its academic teachers, vocational instructors, and educational supervisors, the Office of the Inspector General recommends that the warden, in accordance with and to the extent permissible under his employees’ labor agreements, take the following actions: • Develop a method to hold supervisors in the prison’s education department accountable for ensuring that their employees complete their required daily tasks and obtain prior permission before they leave work early. • Require all supervisors to maintain logs tracking the number of times employees request to leave early and, when necessary, adjust employees’ workloads accordingly. State of California • April 2011 Page 27 • Investigate and take administrative actions against employees who leave work before completing their daily tasks or without prior permission from their supervisors. • Train educational supervisors on how to initiate progressive discipline on employees when appropriate. Bureau of Audits, Office of the Inspector General Page 28 Finding 3 Ineffective Supervisory Oversight and Personnel Practices Concerning Employee Timesheets at Mule Creek State Prison Have Resulted in Costly Mistakes Although the state requires employees to report on their timesheets when they take a full day of time off, we found 14 employees at Mule Creek State Prison (Mule Creek) who failed to claim a total of 23 full days of time off—worth nearly $16,000—between June and August 2010. In nearly all of these 23 cases, the employees and their supervisors signed and approved the timesheets, yet the unreported time off went unnoticed until we brought the errors to the attention of the employees and their supervisors. Employees who fail to report full days of time off will be able to use those hours again in the future, or they will be paid in cash for those hours when they eventually separate from state service. Moreover, during a four-month period between May and August 2010, we found that Mule Creek’s personnel office made many timekeeping mistakes, worth the equivalent of $92,744. Specifically, during this four-month period, the prison either under- or overcharged employees’ leave time on 73 of the 325 timesheets we reviewed (22 percent), belonging to 49 of the 112 employees (44 percent). These errors occurred because the prison’s personnel office failed to correctly add up or properly record the leave hours claimed by the employees. With financial consequences similar to employees’ unreported time off, the personnel office’s non-deducted hours could be used by the employees again or become money that the state would be obligated to pay to the employees upon separation from state service. Conversely, the extra hours deducted from employees, resulting from the overcharging of leave time, essentially cost the employees an extra $6,532. Had we not caught these mistakes, these overcharged employees would not have been able to take the time off that they had earned. In all, as shown in Table 3, we found a total of $108,527 in unreported time off and other timekeeping mistakes. After we brought these errors to Mule Creek’s attention, the personnel Table 3: Summary of Timekeeping Mistakes Number of Timesheets Number of Number of Value of with Reviewed Error Questioned Questioned Description of Mistakes Exceptions Timesheets Rate Hours Hours* Unreported Time off (June through August 2010) 18 115 16% 216 $15,783 (23 Days) Under- and Overcharges to Employee Leave Balances 73 325 22% 1,672 $92,744 (May through August 2010) Total 1,888 $108,527 Source: Mule Creek State Prison’s staff accountability and leave accounting systems, payroll records, and monthly employee timesheets. *The value of questioned hours was calculated by multiplying an employee’s questioned hours by their average hourly salary. We calculated an employee’s average hourly salary by dividing their regular pay for June, July, and August 2010, by the total number of regular hours they were paid for during the same time period. State of California • April 2011 Page 29 office took some initial steps to address some of these issues, including adjusting some of the employees’ leave balances and providing many supervisors with additional timesheet training. However, more improvement is required to fully resolve the problems we found. Several of the prison’s employees failed to report full days of time off on their timesheets, resulting in nearly $16,000 in overpayments The state requires its employees—regardless of whether they are paid a salary or paid by the hour—to report on their timesheet when they take a full day of time off from work. To be compensated for a day off from work, employees must use the equivalent amount of hours from a variety of leave balances, such as vacation, sick, or annual leave. We analyzed whether Mule Creek’s employees neglected to report full days of time off from work for the three months ending August 2010. Toward that end, we used Mule Creek’s accountability system— which identifies the time of day employees entered and exited the secured perimeter—along with other personnel-related data and records. We found that 14 employees failed to report 23 days of time off, at a cost to the prison of nearly $16,000. Four of these 14 employees failed to report multiple days of time off. When we asked the employees about their timesheets, most of these 14 employees indicated that they simply forgot to report their time off on their timesheets. Other employees responded that the questioned days were just regular days off for them; after reviewing these employees’ schedules, however, that does not appear to be the case. One employee no longer works at the prison, so we could not follow up on that case. In 13 of the 14 employees’ cases, the employees and their supervisors signed and approved the timesheets and the unreported time off went unnoticed until we brought it to the employees’ and their supervisor’s attention. The remaining employee’s timesheet was missing the supervisor’s signature altogether. When employees fail to report full days of time off on their timesheets, the employees will be able to use those hours again in the future, or they will be paid in cash for those hours when they eventually separate from state service. When we reconciled the number of hours that Mule Creek employees were paid with the number of hours they spent inside the secured perimeter, in training, or as authorized time off, we found that many employees were paid for hours that we could not account for inside the secured perimeter. We selected 51 of the employees who had a large number of unaccounted- for hours, tested 115 of those employees’ timesheets from a three-month period (which ended August 2010), and compared the days the employees claimed on their timesheets to have worked to the days the employees actually entered the secured perimeter, according to the prison’s accountability system. In rare cases, employees claimed they forgot their identification card for a particular day, which would result in having unaccounted-for hours; in those cases, we used either the prison’s handwritten sign-in roster or other documents instead of the electronic data to verify whether the employee actually showed up to work. In total, we found 42 days during which 19 employees reported on their timesheets that they worked, but the employees never entered the secured perimeter nor attended outside training. We followed up with these employees to discover whether they did, in fact, work, and we found that on 19 of those days, the employees simply switched their regular day off with the day in question but did not indicate the switch on the timesheet. However, we found that 14 employees neglected to report time off on 23 days when they did not show up to work. Bureau of Audits, Office of the Inspector General Page 30 Most of the employees indicated to us that they simply forgot to report their time off and acknowledged they should have done so. However, four of these employees failed to report time off on multiple days. The most egregious case related to an employee who reported on his August 2010 timesheet that he worked on six days at Mule Creek, but he never entered the secured perimeter or attended outside training on those six days. As a result of our follow-up, he indicated to a high-ranking manager that he had inadvertently claimed to work on those days, and he apologized to his manager for doing so. He also said that he would revise his timesheet accordingly. However, we discovered that during four of those days, the employee also reported working at another prison, resulting in his being paid by two prisons for the same four days. In total, the employee should have had six days of time off deducted from his leave balances (60 hours, since he was scheduled to work ten-hour days); but since these days went unnoticed, he was overpaid a total of $7,071 in regular wages for these 60 hours. Although employees are individually responsible for accurate time reporting, supervisors are equally responsible for double-checking their employees’ timesheets. In 22 of these 23 cases of employees taking unreported time off, the supervisors approved and signed the employees’ timesheets while failing to catch the unreported time off; in the remaining case, the employee submitted the timesheet without his or her supervisor’s signature. The supervisors’ inadequate oversight may result in part from ineffective practices of reporting time off at Mule Creek. The prison has a form for employees to fill out when requesting a day off from work and when reporting an unscheduled day off (such as a sick day). Ideally, at the end of the pay period, employees and their supervisors use these forms to ensure that employees claim full days of time off on their timesheets. However, this process can only be effective if employees and supervisors actually use the forms. We attempted to follow up with 13 of the 14 employees or with their supervisors to learn whether any of the employees used the proper form for these 23 unreported days of time off.5 We found that only one employee filled out a form (this accounted for four of the 23 days in question). Yet even in this case, the supervisor, who provided us with a copy of the form, apparently did not look at it when approving the employee’s timesheet. We learned that in 14 of the 23 cases, six supervisors used an alternative method of tracking time off by marking on their calendars or schedules that their employees had taken time off from work; yet again, each supervisor apparently failed to consult his or her alternative tracking mechanisms when approving employees’ timesheets. One of these supervisors told us, “I trust that employees complete their timesheets accurately and I assume my employees have good integrity to complete their timesheets truthfully.” Another supervisor who said he occasionally signs timesheets on behalf of his manager explained that he presumes employees are honest and typically just signs the timesheets. This supervisor also commented that the process is based on the honor system and that he assumes employees worked the days they claim. However, these supervisors miss the point: a superior checks timesheets for accuracy because even honest employees can unintentionally make mistakes. The mistakes we found could have easily been detected if the supervisors had followed Mule Creek’s processes—or even followed their own processes. 5We could not determine whether one employee used a form to request or report time off because the employee separated from employment at Mule Creek State Prison prior to our fieldwork. State of California • April 2011 Page 31 Because of the frequent occurrence of unreported days of time off, we question whether employees and supervisors have been adequately trained to fill out and review timesheets. We asked Mule Creek’s personnel officer how often the prison provides training to its employees on timesheet procedures. According to the personnel officer, new employees typically receive a brief timesheet course as part of the prison’s new employee orientation. However, the personnel officer noted that, to her knowledge, Mule Creek had not conducted any other formal courses focusing on timekeeping until we brought several of our concerns about timekeeping to the prison’s attention. Twice in December 2010, and once in January 2011, the personnel office provided a total of 54 supervisors with timesheet training, which included handouts of various rules as well as helpful examples of erroneous timesheets that had been recently completed and approved. Although this appears to be a good start on addressing the problem, Mule Creek must continue to emphasize the importance of accurate time reporting and include additional employees in the training to ensure more accurate reporting. Inaccuracies like the ones reported here have the potential to be very costly to the prison. Timekeeping mistakes made by the prison’s personnel office resulted in dozens of employees receiving extra time-off hours Employees at Mule Creek’s personnel office often deducted too many or too few leave hours as they reviewed employee timesheets and posted their leave hours into the state’s leave accounting system. After we performed our data analysis (as described on page 11), we selected 112 employees primarily from classifications having a large number of unaccounted-for hours and reviewed 325 of their timesheets (from selected months between May and August 2010). Table 4 on the following page shows a summary of the classifications of employees and their timesheets reviewed. We found that Mule Creek either under- or overcharged leave time on 73 of the 325 timesheets we reviewed (22 percent) belonging to 49 of the 112 employees (44 percent). These mistakes occurred because the prison’s personnel office failed to add up or properly record the correct number of claimed leave hours. Of the 73 mistakes, we found 55 instances of employees being undercharged, at a total of 1,582 leave hours, and 18 instances of employees being overcharged, at a total of 90 leave hours. At the compensation rates of these employees, these mistakes equated to $86,212 worth of leave that the employees were undercharged. This represents hours of time off that could be used again by the employees or that the employees could receive in cash when they separate from state service. Conversely, the overcharged employees lost $6,532 in time off because of the processing errors. Combined, these errors cost the prison and the employees a total of $92,744 as shown in Table 5 on page 34. Most of the mistakes described above were caused by illegible writing or by common arithmetic errors; however, some mistakes resulted from Mule Creek’s inconsistent application of the state’s rules for using or accruing holiday time-off and excess hours.6 The mistakes attributable to the mismanagement of these rules totaled $2,202. 6Excess hours are accrued or taken to balance an employee’s pay period hours when an employee works a schedule other than Monday through Friday, eight hours a day, to total 40 hours per week. If an employee has worked more than the number of days or hours required for the pay period, the extra time is credited as “excess hours.” For an employee who has worked less than the number of days or hours required, the difference would first be deducted from an employee’s accrued excess hours, then from the employee’s other available leave balances. Bureau of Audits, Office of the Inspector General Page 32 Table 4: Summary of the Employees and Timesheets We Reviewed (May through August 2010) Chief Executive Officer’s Number of Employees Number of Timesheets (CEO) Employees Reviewed With Exceptions Reviewed With Exceptions Psychologists 29 14 86 22 Dental Assistants 13 1 35 1 Psychiatrists 13 10 46 18 Psychiatric Technicians 12 11 33 16 Clinical Social Workers† 8 5 26 7 Physicians and Assistants 6 2 14 2 Nursing Staff 5 2 13 3 Pharmacy Staff 5 0 11 0 Dentists 2 0 6 0 CEO’s Administrative Staff 2 0 6 0 Recreational Therapists 2 0 6 0 CEO’s Employee Totals 97 45 282 69 Warden’s Employees Maintenance Support Staff 5 1 15 1 Correctional Officers 3 0 8 0 Correctional Sergeants 2 1 6 1 Education and Vocation Staff 2 1 5 1 Chaplains 1 1 3 1 Food Preparation Staff 1 0 3 0 Warden’s Administrative Staff 1 0 3 0 Warden’s Employee Totals 15 4 43 4 Totals 112 49 325 73 Source: Mule Creek State Prison’s monthly employee timesheets. † We included one retired annuitant that we did not include in our other analyses. Mistakes resulted in employees being over- and undercharged for their leave taken We compared the total number of leave hours reported by the employees on their timesheets with the total number of leave hours that the timekeeper posted into the California Leave Accounting System (CLAS), the system that captures state employee leave transactions and balances. On 55 of the 325 timesheets, Mule Creek’s personnel office failed to record all of the employees’ claimed leave hours properly, resulting in 1,582 hours undercharged to employees’ leave balances. On 24 of the 55 timesheets, the prison did not record any leave hours at all, even though the employees’ timesheets clearly indicated that they did not work on particular days during the pay period. This is significant because in these instances, employees attempted to use leave hours, but their timesheet declarations passed unrecorded by the prison’s timekeeping process. In the most egregious example, our review of an employee’s timesheet revealed that the employee had declared taking 160 hours of leave time during the June pay period. However, according to CLAS for the same period, the timekeeper recorded zero hours of leave time—giving the employee an “extra” 160 hours of leave time in his or her accrued balances. In response to these mistakes, Mule Creek’s personnel officer is working to correct State of California • April 2011 Page 33 Table 5: Summary of Mistakes Made by the Prison’s Personnel Office When Recording Employee Leave (May through August 2010). Number of Number of Value of Timesheets With Questioned Questioned Description of Mistakes Exceptions Error Rate* Hours Hours† Undercharges to Employee Leave Balances - 55 17% 1,582 $86,212 (where the prison failed to deduct enough leave hours from the employee) Overcharges to Employee Leave Balances - 18 5% 90 $6,532 (where the prison deducted too much leave from the employee) Total Mistakes 73 22% 1,672 $92,744 Source: Mule Creek State Prison’s monthly employee timesheets. * We reviewed a total of 325 timesheets between May and August 2010. † The value of questioned hours was calculated by multiplying an employee’s questioned hours by their average hourly salary. We calculated an employee’s average hourly salary by dividing their regular pay for June, July, and August 2010, by the total number of regular hours they were paid for during the same time period. the affected employees’ leave balances. Had we not brought these mistakes to Mule Creek’s attention, the affected employees would have been able to use the hours again or would have been paid cash for those hours when they retired or otherwise separated from state service. In the remaining 31 instances of undercharged employee leave, Mule Creek’s personnel office recorded some portion of the employees’ claimed time off; however, the prison did not record the full number of leave hours. For example, one employee’s timesheet for the July 2010 pay period showed that the employee took 106 hours of time off. However, according to CLAS records we reviewed, the timekeeper only recorded 42 of those 106 hours—which left the employee with an “extra” 64 hours of leave time. Conversely, in 18 instances, Mule Creek’s personnel office incorrectly recorded more leave hours than employees had declared on their timesheets. For example, in one instance, we found that an employee’s timesheet reflected 56 leave hours taken for June 2010, yet the prison’s personnel staff recorded in CLAS a total of 75 leave hours, thereby overcharging the employee 19 leave hours (or $2,305 at this employee’s compensation rate). In 17 other instances, the prison similarly took too many leave hours from employees. In total, the prison overcharged employees 90 hours of leave, worth about $6,532. Had we not brought these mistakes to Mule Creek’s attention, these overcharged employees would not have been able to take the time off that they had earned. Mule Creek inconsistently accounts for holiday time and inconsistently balances pay period hours Contributing to the leave-time mistakes we found, Mule Creek miscalculated holiday time and other time credits for employees who worked alternative work schedules.7 When a state holiday is observed, full-time employees are allotted eight hours of paid leave. Generally speaking, when an employee who is scheduled to work a nine-hour or ten-hour shift takes a 7An alternative work schedule is an “alternative” to the traditional schedule of eight-hour workdays, Monday through Friday, equaling a 40-hour workweek. Alternative work schedules may result in the employee working more or fewer hours in a pay period. Bureau of Audits, Office of the Inspector General Page 34 holiday off, the employee must contribute additional leave hours to make up the difference between the eight-hour paid holiday leave and his or her nine-hour or ten-hour scheduled day. In addition, employees on alternative work schedules who work more or fewer than the number of days or hours required for any pay period must have hours added to or subtracted from an “excess hour” account or other leave accounts. The mistakes we found mostly benefitted the employees, but sometimes the prison overcharged hours from employees’ leave or “excess” hour accounts. Overall, the net effect of these errors was $2,202. Of the 325 timesheets we reviewed, 37 related to employees who worked nine- or ten-hour days. Of those 37 timesheets, we found 17 instances (46 percent) in which Mule Creek failed to take the additional one or two leave hours on the Memorial Day and Independence Day holidays. Alternatively, we found three instances in which the prison failed to credit holiday hours to employees working a weekly four-day, ten-hour schedule when their regular day off fell on a holiday. Instead, Mule Creek mistakenly charged the employees two additional hours of leave, as if the holiday were observed on the employees’ scheduled workday. In addition to the errors involving holiday hours, we identified six errors relating to leave hours that certain employees on alternative work schedules occasionally earned or expended to balance the pay period. In four instances, we found that Mule Creek failed to charge the employees leave hours when the employees worked fewer than the required monthly pay- period hours. Conversely, in two separate cases, the prison overcharged leave to employees who were not supposed to earn excess leave hours. The Memorandum of Understanding for these employees does not require the documentation of hours worked; therefore, these two employees should not have been charged leave to balance the pay period. Several factors contributed to the mistakes We met with the prison’s warden and personnel officer to discuss the problems we found, obtain their perspective, and learn of any corrective action they might have already started. Generally, they speculated that the errors were caused by four factors: poor supervisory oversight, poor training, heavy workloads for the timekeeper, and poor communication between the personnel office and the warden. The warden told us that he became aware of timekeeping mistakes shortly after his appointment in 2008, and that he has worked to improve the quality of timekeeping at Mule Creek. He said that he acted initially to eliminate the cause of the mistakes but that his efforts were unsuccessful. The warden noted that he eventually discovered the problems to be widespread and that a lack of information hindered his full insight into the timekeeping matter. Since February 2010, a new personnel officer has overseen the timekeeping function. The warden has tasked this new person with evaluating and reporting on current timekeeping deficiencies at Mule Creek. The personnel officer told us that although supervisors are responsible for reviewing and approving employee’s timesheets, the supervisors often do not closely track employees’ schedules and, therefore, often lack sufficient information to ensure that the leave time actually taken is accurately shown on the employee’s timesheet. The personnel officer added that State of California • April 2011 Page 35 supervisors also do not enforce a policy from the California Department of Corrections and Rehabilitation that requires employees to add up their various hours on their timesheets, and that moreover, supervisors are approving timesheets with illegible writing. According to the personnel officer, these problems place a burden on the already overworked timekeepers as they try to read illegible writing and total the hours worked and leave taken. Finally, the personnel officer told us that there is usually only one timekeeper performing data entry into CLAS for Mule Creek’s approximately 1,300 employees, and that this timekeeper has about ten working days each month to read and post the timesheet information from those 1,300 timesheets. During that ten-day period, the timekeeper also answers the telephones and greets employee-visitors at the front counter—added distractions that may contribute to mistakes. Additionally, the timekeeper was new to the job in May 2010. Recommendations To ensure that all employees understand the importance and methods of accurate timekeeping, the Office of the Inspector General recommends that the warden take the following actions: • Provide training to all employees on how to properly complete timesheets. This training should include an emphasis on the importance of totaling the various types of leave hours charged each month and on writing legibly. • Provide training to all supervisors and managers on how to properly review completed timesheets. This training should include an emphasis on ensuring that employee leave hours are accurate and on returning incomplete or inaccurate timesheets to employees for corrections. • Require all employees to use the prison’s form for requesting and reporting time off when taking time off from work for any reason. • Study whether Mule Creek has a sufficient number of timekeepers. If the prison lacks a sufficient number of timekeepers, redirect employees in the Personnel Office to assist the existing timekeepers. • Instruct timekeepers to reject incomplete or illegible timesheets. Where appropriate, provide remedial instruction to employees or supervisors who persist in not following existing policies related to timesheets. • Provide training to timekeepers, payroll specialists, and payroll specialist supervisors on the correct method of processing and reviewing employees’ monthly leave activity and leave balances in the California Leave Accounting System. Bureau of Audits, Office of the Inspector General Page 36 Appendix In the table on the following page, we present the detailed calculations of unaccounted-for hours—hours for which employees were paid but which we could not reconcile with the employees’ documented time inside the prison’s secured perimeter, their time off, and their time spent in training. These are the unaccounted-for hours of Mule Creek State Prison’s psychiatrists, psychologists, licensed clinical social workers, academic teachers, vocational instructors, and educational supervisors for the period of June through August 2010. For this period, we computed over 4,000 unaccounted-for hours for these employees at a cost of $272,900 (or $1,091,600 annualized). To arrive at an employee’s unaccounted-for hours, we began with the employee’s total number of paid hours; then we subtracted from that the number the hours the employee spent inside the secured perimeter, the hours the employee took as time off, and the hours the employee spent in training. We considered any results greater than zero to be “unaccounted-for” because the employee was paid for more hours than we could account for using known data fields. Our analysis suggests that unaccounted-for hours stem mainly from employees’ not averaging a full day inside the secured perimeter; yet the unaccounted-for hours can also result from a combination of other causes, including the following: timekeeping mistakes made by Mule Creek’s personnel office when entering declared time off into the state’s leave accounting system; employees working a small portion of their day outside of the secured perimeter; and employees forgetting to report full days of time off on their timesheets. In addition, on rare occasions, employees’ unaccounted-for hours resulted from their forgetting their identification cards on a particular day and having to manually sign in and out of the prison. Without an identification card, an employee would not be able to scan his or her entry and exit into the prison’s Staff Accountability System. Six employees in these classifications actually had negative variances because we could account for more hours than they were paid during the period. However, for the purposes of our review, Table A on the following page presents these employees as having zero unaccounted- for hours so that their negative variances do not offset other employees’ shortages of hours. State of California • April 2011 Page 37 Table A: Summary of Mental Health and Education Employees’ Unaccounted-for Hours (June through August 2010) Hours -( Hours Inside + Hours of + Training )= Unaccounted-for Cost of Employee Paid the Secured Authorized Hours Hours Unaccounted-for Perimeter Time Off Hours* Mental Health Employees Psychiatrist 1 528 230 178 0 120 $14,129 Psychiatrist 2 528 297 98 3 130 15,201 Psychiatrist 3 528 307 70 1 150 17,629 Psychiatrist 4 112 72 8 0 32 3,085 Psychiatrist 5 264 191 0 0 73 8,869 Psychiatrist 6 528 352 78 0 98 11,608 Psychiatrist 7 528 237 58 1 232 27,114 Psychiatrist 8 528 353 84 39 52 6,297 Psychiatrist 9§ 528 118 263 2 145 17,617 Psychiatrist 10 528 282 174 1 71 8,610 Psychiatrist 11 528 251 178 9 90 10,564 Psychiatrist 12§ 528 480 40 2 6 662 Psychiatrist 13 176 166 8 9 0 0† Subtotals 5,656 3,170 1,229 58 1,199 $141,385 Psychologist 1 240 24 146 0 70 $3,502 Psychologist 2 528 243 180 43 62 2,952 Psychologist 3 528 384 88 1 55 2,517 Psychologist 4 448 315 10 6 117 5,873 Psychologist 5 528 384 70 12 62 2,990 Psychologist 6 528 309 132 0 87 4,171 Psychologist 7§ 528 363 150 8 7 339 Psychologist 8 528 364 80 14 70 3,240 Psychologist 9§ 528 415 108 0 5 250 Psychologist 10 528 397 78 7 46 2,232 Psychologist 11 528 337 132 6 53 2,520 Psychologist 12 530 409 68 49 4 182 Psychologist 13 352 73 250 0 29 1,344 Psychologist 14 528 359 40 1 128 6,188 Psychologist 15 528 331 68 1 128 6,202 Psychologist 16 528 447 0 1 80 3,860 Psychologist 17 528 347 112 8 61 2,943 Psychologist 18 528 341 80 9 98 4,726 Psychologist 19 528 318 126 0 84 4,070 Psychologist 20 528 413 20 0 95 4,530 Psychologist 21 528 404 40 2 82 3,982 Psychologist 22§ 528 443 8 7 70 4,269 Psychologist 23 528 399 31 9 89 4,293 Psychologist 24 528 451 50 5 22 985 Psychologist 25§ 528 421 40 42 25 1,229 Psychologist 26 528 418 70 1 39 1,851 Psychologist 27§ 528 481 96 3 0 0† Psychologist 28 528 442 110 44 0 0† Psychologist 29 176 150 48 1 0 0† Psychologist 30 528 453 100 6 0 0† Psychologist 31§ 528 537 89 6 0 0† Subtotals 13,186 9,109 2,177 232 1,668 $81,240 Continued Bureau of Audits, Office of the Inspector General Page 38 Table A: Summary of Mental Health and Education Employees’ Unaccounted-for Hours (June through August 2010) - continued Hours -( Hours Inside + Hours of + Training )= Unaccounted-for Cost of Employee Paid the Secured Authorized Hours Hours Unaccounted-for Perimeter Time Off Hours* Licensed Clinical Social Worker 1 528 329 118 1 80 $2,734 Licensed Clinical Social Worker 2 528 407 18 52 51 1,995 Licensed Clinical Social Worker 3 528 465 20 13 30 1,169 Licensed Clinical Social Worker 4 528 352 96 7 73 2,878 Licensed Clinical Social Worker 5 528 381 64 0 83 2,876 Licensed Clinical Social Worker 6 454 313 50 1 90 3,686 Licensed Clinical Social Worker 7 528 296 110 10 112 4,426 Subtotals 3,622 2,543 476 84 519 $19,764 Mental Health Totals 22,464 14,822 3,882 374 3,386 $242,389 Education Employees Teacher 1 528 277 192 0 59 $2,218 Teacher 2 528 193 272 2 61 2,464 Teacher 3 528 342 136 3 47 1,655 Teacher 4 528 324 168 3 33 958 Teacher 5 528 349 144 3 32 1,251 Teacher 6 528 336 160 0 32 1,134 Teacher 7 528 378 128 3 19 769 Teacher 8 528 348 160 2 18 599 Teacher 9 528 341 176 0 11 433 Teacher 10 528 352 168 3 5 211 Teacher 11 528 383 120 3 22 864 Teacher 12 528 376 144 3 5 162 Subtotals 6,336 3,999 1,968 25 344 $12,718 Vocational Instructor 1 528 298 176 0 54 $2,036 Vocational Instructor 2 528 353 136 3 36 1,256 Vocational Instructor 3 528 332 144 3 49 1,770 Vocational Instructor 4 528 348 152 0 28 995 Vocational Instructor 5 528 343 144 3 38 1,433 Subtotals 2,640 1,674 752 9 205 $7,490 Education Supervisor 1§ 528 382 40 3 103 $4,592 Education Supervisor 2§ 528 390 64 2 72 2,916 Education Supervisor 3§ 528 298 160 1 69 2,795 Subtotals 1,584 1,070 264 6 244 $10,303 Education Totals 10,560 6,743 2,984 40 793 $30,511 Mule Creek Totals 33,024 21,565 6,866 414 4,179 $272,900 Annualized Totals 16,716 $1,091,600 Source: Mule Creek State Prison’s staff accountability, in-service training, and leave accounting systems and payroll records (June through August 2010). * We calculated the cost of unaccounted-for hours by multiplying an employee’s unaccounted-for hours by their average hourly salary. We calculated an employee’s average hourly salary by dividing their regular pay for June, July, and August 2010, by the total number of regular hours they were paid for during the same period. § These employees are supervisors. † We accounted for all of these employees’ hours. Therefore, we exclude these employees from the totals so that their extra hours worked would not balance out other employees’ unaccounted-for hours. State of California • April 2011 Page 39 California Department of Corrections and Rehabilitation’s response to the special report (page 1 of 2) Bureau of Audits, Office of the Inspector General Page 40 California Department of Corrections and Rehabilitation’s response to the special report (page 2 of 2) State of California • April 2011 Page 41 SPECIAL REPORT Spe IN c MA i T a E C l EL L R P e HO p NE o US R E E t NDANGERS PRISON SECURITY AND PUBLIC SAFETY Mule Creek State priSon MuSt iMprove itS overSight of SoMe OFFICE OF THE eMployeeS’ Wor k hourS and tiMekeeping INSPECTOR GENERAL DAVID R. SHAW office of the inSpectINoSPREC TgORe GEnNEeRARL al STATE OF CALIFORNIA MAY 2009 Bruce a. Monfross inSpeCtor general (a)  William Shepherd deputy inSpeCtor general, in-Charge Bryan Beyer deputy inSpeCtor general, Senior Christina animo deputy inSpeCtor general ryan Baer deputy inSpeCtor general kristine rodrigues deputy inSpeCtor general Ben Ward deputy inSpeCtor general Bert Wenzel deputy inSpeCtor general State of califoRnia april 2011 WWW.oig.Ca.gov